Voice detection

Abstract

In a voice processing system, a voice detection algorithm is disclosed wherein the sensitivity is adjusted based upon whether or not tones (e.g.; DTMF) are presently being received. In an additional embodiment, the adjustment is made only when the voice processing system is in a predetermined mode.

We claim: 
     
       1. In a voice processing system, a method of detecting speech and predetermined tones in a signal being received from a telephone network, said method comprising the steps of: detecting whether or not at least one of said predetermined tones is present in said signal being received from said telephone network and indicating if said at least one of said predetermined tones is present; and   preventing speech from being detected, in response to said step of detecting indicating the presence of at least one of said predetermined tones.   
     
     
       2. The method of claim 1 wherein said step of preventing occurs only when said step of detecting detects a valid tone and when said voice processing system is performing a predetermined one of a plurality of functions. 
     
     
       3. The method of claim 2 wherein said step of preventing includes the step of holding an input to a logic gate at a logical zero.